Index analysis


The following analysis is based on IS questions in the LG21 staff survey, using only complete responses
(10,089). The original survey response data have been transformed, so every staff response is assgined an
Index score between 1 (most negative responses) and 5 (most positive responses). Only selected questions
contributed to the Index for each topic (see table of contents for index questions). The boxplots below show
Index scores for IS questions for each Respondent (Fig A - boxplots are standardized graphs displaying
data distributions based on a five number summary [minimum, first quartile (Q1), median, third quartile (Q3),
and maximum]. Overall, IS questions scored lower other topics (i.e. scores are significantly
lower than other topics, Fig A). Within the IS Index scores, SOAs gave more negative feedback
than other Roles, contributing to their lower Index.


Fig A). LG21 Survey Responses, aggregated by all Topics (left), and by Role for IS (right)


Then we can break the IS Index scores down by staff demographics, and the geographical remoteness of the Venue where individual staff worked (Fig B). IS Index scores did not differ by demographics (do it), indicating the background of LG21 Election Staff did not substantailly affect their survey responses (how to phrase?). Similary, the IS Index scores where similar for staff working in the major cities or regional towns.


Fig B). IS Survey Responses by Demoghraphic (left) and remoteness (right)


Index Map

Fig C). Vote IS Survey Responses by RO Region


The map below shows the Index of survey reponses for IS questions in the LG21 Staff Survey. The index
values are the same as the boxplots, but are the average of all the staff responses within each RO Region
(the blue boundaries on the map). The blank Regions are those that did not participate in LG21 (Western,
Central Coast, Penrith and Wingecarribee LGAs).